There are a large number of partly contradictory optimization objectives when constructing and designing photovoltaic systems for power generation. In order to provide a “good” photovoltaic system, a satisfactory compromise from various optimization objectives must be found.
However, existing software tools for the optimization of PV systems and the underlying concepts and methods follow a one-objective approach, as a result of which, by principle, they only assist the user of the software tool insufficiently with the configuration of the PV system. Furthermore, such approaches tend to divide optimization scope into individual criteria.
The basic principle of power generation in a photovoltaic system can be described as follows: photovoltaic modules convert incoming sunlight into direct current. Using “inverters” (that is to say DC-AC converters in the sense of power inverters), this direct current is converted into an alternating current (of the local grid frequency, that is to say 50 Hz or 60 Hz) with higher voltage, and the current thus produced is fed after a further voltage increase by transformers into the power grid of an energy supplier.
Here, the PV modules are not usually assembled individually in the developed area, but a relatively large number of modules are combined to form a relatively large module, or what is known as a “panel”. A panel stands on a number of feet and for example can carry 100 modules, which are mounted on the panel in a number of rows, for example five rows, for example so as to form twenty modules in each case. The longer is such a panel, the more supporting feet it has, which can also be held in the transverse direction with struts in the manner of a base frame, or can carry the solar modules in the manner of 3D grid constructions.
Since the voltage supplied by an individual module is too small to be fed directly into an inverter, a number of modules are connected in series to form what are known as strings. By way of example, one solar panel would contain five strings each formed by twenty modules electrically connected in series, wherein the five strings could correspond to the five module rows of the panel. The strings of a panel are connected in parallel in the example. Lastly, a number of solar panels are connected in parallel to an inverter input (for example as a power inverter). Other electrical connections of the modules, for example in the form of a butterfly layout, are also possible. Here, two strings share two adjacent module rows in each case. Preferred voltage ranges of the multiplied DC voltage (multiplied by the number of modules per string) obtained in this way may lie above 500 V, preferably in the range between 700 V and 1,500 V.
For the purpose of improved utilization of the incoming solar radiation, the modules may be mounted not flat (horizontally) on the panel, but inclined by a certain angle of inclination from the horizontal in the direction of the equator, as can be achieved by different lengths of the feet or of the base frame.
The above-mentioned weak points of existing PV system optimization tools, that is to say both the insufficient decision support in the search for a compromise and also the lack of optimization scope, are to be remedied/overcome as (technical) problems.